Summary Identifying gene function and impact on disease biology are overarching aims of life science research in the post-genomic era. Functional genomics also underpins our ability to understand the meaning of genetic variation in human populations. However, crucial gaps remain in the functional genomics tool set that will slow our progress in applying genomics to unravel disease biology. Currently, efficient pooled methods for genome-wide screening require either selection of cells based on growth advantage or physical purification (eg by FACS). Many disease processes are characterized by complex cellular phenotypes including defects in cell or organelle morphology, subcellular localization of molecular components, or cell motility. Other key phenotypes of interest may involve transient states (eg mitosis), cell-cell interactions, or require dynamic assays in live cells (eg, optical recording of electrophysiological activity of cardiac or neural cells). Image- based, high-content screens using overexpression and RNA interference have uncovered novel genes involved in complex phenotypes, including mitosis, synaptogenesis, and embryogenesis. However, such microplate-based screens of clonal cell populations are not regularly conducted at the genomic scale due to the expense, labor and automation expertise required. Although ?living cell array? screens have reduced some logistical hurdles, they still require individually synthesizing and arraying each gene perturbation reagent. Other possible approaches for pooled screening such as single-cell transcriptomics cannot access the range of complex and dynamic disease-associated phenotypes needed. Here we propose to develop a new genomic perturbation and screening concept that combines major advantages of pooled perturbation with imaging assays for single-cell arrayed readout of complex phenotypes. Specifically, we will screen pooled genomic perturbations (with barcoded CRISPR-Cas9 single- guide RNAs) using microscopy to read out phenotypes AND to identify perturbed genes at the single-cell level. Perturbed genes will be identified by sgRNA-associated expressed barcodes in situ by read RNA fluorescence in situ hybridization (FISH) or in situ sequencing (IS). This approach is highly scalable because the limiting cost is microscope time for imaging. We image millions of cells per day on a standard research microscope and up to 100 million cells per day on a high-throughput screening microscope. This means that genome-wide screens can be routinely conducted on a general-purpose microscope in small labs. Even more ambitious screens such as all pairwise combinations of 1000 genes with each pair represented across hundreds of cells (? 108 single-cell assays in toto) could be carried out routinely in a dedicated facility.